Current information, on demand
Web Search is $10 per 1,000 calls. Sources return with the answer, not in a separate research tab.
One account. The right model every time.
Frontier AI without the provider maze.
Chat, search, run code and reuse context without managing a stack of provider accounts.
Itzi Gateway
One stable surface for models, tools, caching and fallback routing.
- 13
- customer models
- 3
- compatible API formats
- 5 min
- customer cache window
- $0.50
- welcome credit
Keep the context. Stop paying full price for it.
Auto Cache detects reusable conversation prefixes on supported OpenAI and Anthropic models. You control a clear five-minute cache window.
Full reusable conversation prefixAutomatic API cache placementSeparate read and write pricing
System and instructionsretained
Conversation historyretained
Latest turnnew
Without cache
Full input
With Auto Cache
Reuse the prefix
More than a model picker.
Itzi gives every supported model the tools needed to finish real work inside one conversation.
Python Studio
Run analysis and code in an isolated environment for $5 per 1,000 runs.
Your own tools
Send standard function definitions and let compatible models call your application directly.
Automatic caching
Auto Cache finds the right reusable prefix while explicit cache controls remain available.
Usage you can inspect
Every request shows tokens, cache activity, tool charges and the customer price paid.
Choose the model, not the account.
Customer prices stay identical across chat and API. Cache rates appear only when the model supports them.
| Model | Input | Output | Cache | Context | Capabilities |
|---|---|---|---|---|---|
Claude Opus 4.8 claude-opus-4.8 | $5.00 | $25.00 | Read $0.75 Write $10.00 | 1M | VisionToolsSearchAuto Cache |
Claude Opus 5 claude-opus-5 | $5.00 | $25.00 | Read $0.75 Write $10.00 | 1M | VisionToolsSearchAuto Cache |
GPT 5.6 Terra gpt-5.6-terra | $2.50 | $15.00 | Read $0.38 Write $5.00 | 248K | VisionToolsSearchAuto Cache |
GPT 5.6 Sol gpt-5.6-sol | $5.00 | $30.00 | Read $0.75 Write $10.00 | 248K | VisionToolsSearchAuto Cache |
Gemini 3.6 FlashBeta gemini-3.6-flash-thinking | $1.50 | $7.50 | Not supported | 1M | VisionToolsSearch |
Kimi K3 kimi-k3 | $3.00 | $15.00 | Not supported | 1M | VisionTools |
Prices are shown in USD per million tokens.
Full catalogPlans that refill
Subscribe for usage, not feature gates.
Every plan unlocks the same catalog and tools. Your included allowance refreshes every seven days.
Compare every planMost popular
Plus
$19per month
For longer sessions and regular use across the model catalog.
$22
of usage credit every seven days
Keep the SDK you already use.
Point OpenAI, Anthropic or Responses-compatible clients at Itzi. Chat and API share the same balance and pricing.
Read the API overviewOpenAI-compatible request
import OpenAI from "openai";
const client = new OpenAI({
apiKey: process.env.ITZI_API_KEY,
baseURL: "https://itzi.app/v1",
});
const response = await client.chat.completions.create({
model: "claude-opus-4.8",
messages: [{ role: "user", content: "Review this proposal" }],
// Auto Cache is enabled automatically on supported models.
});Start with real credit.
Create an account and use the included $0.50 across chat, API, caching and tools.